---
title: "Key-value pair extraction solution | Nutrient .NET SDK"
canonical_url: "https://www.nutrient.io/guides/dotnet/extraction/key-value-pairs/kvp-overview/"
md_url: "https://www.nutrient.io/guides/dotnet/extraction/key-value-pairs/kvp-overview.md"
last_updated: "2026-05-20T19:49:34.791Z"
description: "Automatically extract key-value pairs from documents using AI and ML with our .NET SDK for C# applications."
---

# Extract key-value pairs with our .NET SDK

Automatically detect and extract key-value pairs from unstructured documents or images with our native SDK. Leveraging AI, ML, and adaptive layout understanding, the key-value pair (KVP) engine is used to reduce manual and repetitive document processing tasks. It can automatically label and extract information such as phone numbers, IBANs, credit cards, names, and email addresses from unstructured or semi-structured documents. The engine recognizes KVPs in all supported OCR languages.

As a native SDK, it can be deployed on-premises or embedded in your application, and it works offline, without internet access. Integration is simple and doesn’t require specific types of CPUs or a complex server infrastructure. Explore our guides and code samples to learn how you can integrate our KVP technology into your C# application or workflow.

[Try for Free](https://www.nutrient.io/sdk/dotnet/getting-started.md)

[Launch Demo](https://www.nutrient.io/demo/data-extraction)

## Key capabilities

- Powered by AI and ML — 15+ years of continuous improvements in accuracy

- Automatically detects key values and creates labels for data sets: phone numbers, IBANs, names, credit card numbers, and more

- Leverages AI, ML, heuristics, mathematics, and adaptive layout understanding to identify key values

- Works on all document types, including unstructured and semi-structured documents or documents without predefined data models

- Integration into C# applications with simple CPU or server architecture requirements

- 100 percent control over your document security and privacy via the native SDK

## Guides for key-value pair extraction

**[How key-value pair extraction works](https://www.nutrient.io/guides/dotnet/extraction/key-value-pairs.md)**

Learn how to use our key-value pair engine

**[Data model](https://www.nutrient.io/guides/dotnet/extraction/key-value-pairs/data-model.md)**

Learn about the data model behind the extraction technology

**[Confidence score](https://www.nutrient.io/guides/dotnet/extraction/key-value-pairs/confidence-score.md)**

Learn about how confidence scores are determined

**[Data types](https://www.nutrient.io/guides/dotnet/extraction/key-value-pairs/data-types.md)**

Learn about automatically detected data types

**[Extract data from bank statements](https://www.nutrient.io/guides/dotnet/extraction/bank-statements.md)**

Learn how to extract data from bank statements

**[Extract data from tables](https://www.nutrient.io/guides/dotnet/extraction/tables.md)**

Learn how to extract data from tables

**[Extract data from invoices](https://www.nutrient.io/guides/dotnet/extraction/invoices.md)**

Learn how to extract data from invoices
---

## Related pages

- [Key-value pair extraction data types](/guides/dotnet/extraction/key-value-pairs/data-types.md)
- [How key-value pair extraction works](/guides/dotnet/extraction/key-value-pairs.md)
- [Understand confidence scores in KVP extraction](/guides/dotnet/extraction/key-value-pairs/confidence-score.md)
- [Key-value pair extraction methods](/guides/dotnet/extraction/key-value-pairs/data-model.md)
- [Read key-value pair data from an image](/guides/dotnet/extraction/key-value-pairs/read-kvp-data-from-image.md)